Analyzes your `.agent.md` and `.prompt.md` files to recommend the best AI model from GitHub Copilot's lineup based on task complexity and your subscription tier. It categorizes your prompts into types like simple refactoring, complex architecture work, or debugging, then maps them against model capabilities like reasoning depth, code quality, and cost multipliers. Handles the full model matrix from GPT-5 mini for quick tasks to Claude Opus 4.1 for large codebase analysis. Factors in context window needs, vision requirements, and whether you're on Free, Pro, or Pro+ plans. Saves you from burning premium credits on tasks that could run fine on 0x multiplier models.
npx skills add https://github.com/github/awesome-copilot --skill model-recommendation